Lloyds Bank Plc v Mrs M Fox & Ors [1992] UKEAT 556_89_3007

EMPOLYMENT APPEAL TRIBUNAL
BAILII case number: [1992] UKEAT 556_89_3007Case No Appeal No. EAT/556/89Venue 58 VICTORIA EMBANKMENT, LONDON EC4Y 0DS
Between
Lloyds Bank PlcMrs M Fox & Ors
Date 30 July 1992
MR JUSTICE WOOD (PRESIDENT): This Appeal Tribunal has been unable to reach a unanimous decision. The majority (Ms Smith dissenting) will allow the appeal. Ms Smith finds herself unable to accept the reasoning of the majority and her own reasons for dismissing the appeal are set out separately towards the end of this judgment.Of the Applicants, Mrs Longman is a clerk-typists and the remainder each of them secretaries and typists in the employ of Lloyds Bank Plc ("the Bank"). They each claim that a term in their contract of employment should be altered under the provisions of the Equal Pay Act 1970, so that they receive the same pay as Mr Martin - a senior messenger. For the purposes of these claims only, the Bank has conceded that the work of the employees is of equal value with the work of Mr Martin. The issue raised therefore, is under S.1(3) of the 1970 Act - the existence of a genuine material factor "which is not the difference of sex".

The defence of the Bank is pleaded thus:-

"At all material times, the rates of pay for the work of secretaries and typists (the First to Sixth Applicants' work), the work of clerk-typists (the Seventh Applicant's work) and the work of messengers (the comparators' work) have been determined by separate collective bargaining and arbitration awards. Since the coming into force of the 1970 Act, the different rates of pay have been determined without regard to and irrespective of the sex of the employees."
In their reply the Applicants raise both direct and indirect discrimination. We need only give their plea on direct sex discrimination. It reads:-
"At all material times, the respondents were and are directly discriminating against the applicants on the ground of their sex, in that, if secretaries and/or typists, and/or clerk-typists, employed by them did not consist wholly or mainly of women, and/or if messengers employed by them did not consist wholly or mainly of men, then higher rates of pay would be awarded to the applicants in each of them."
Thereafter the Applicants set out a number of particulars to which reference can if necessary be made. We would only comment that many of those particulars seem more apposite to an allegation of indirect rather than direct discrimination.The Industrial Tribunal found in favour of the Applicants on the basis of direct discrimination, but against them on the issue of indirect discrimination. The Bank appeals; the Applicants cross-appeal. The law concerning the latter is awaiting clarification from the European Court of Justice in ENDERBY v. FRENCHAY HEALTH AUTHORITY . By consent the cross-appeal is therefore dismissed. We should state, merely for the purposes of clarity, that in so far as any decisions were referred to in arguments before us which were binding upon us, in those cases each party reserved the right to argue the matter afresh before any higher court should this appeal go further.At the risk of being repetitive, we stress that the sole issue before us in this appeal relates to direct discrimination. We are not dealing with those factors to be found within the provisions of S.(1)(b) of the Sex Discrimination Act 1975 which are highly material when considering indirect discrimination.

History of the Litigation

The Originating Applications are each of them dated 14th February 1986. This is therefore the material date upon which each applicant's claim is to be judged. The history of this case shows once again the complication of the procedures which at present exist and the scope for tactical manoeuvring.One of the problems has been that the law has been in a state of uncertainty. Notices of Appearance were filed by the Bank in April 1986 and in August 1986 and again in December of that year amendments were made to rely upon PICKSTONE v. FREEMANS PLC [1986] ICR 886, AND [1988] ICR 697 : AND LEVERTON v. CLWYD COUNTY COUNCIL [1987] ICR 158 and [1989] ICR 33. Neither of these amendments was effective as a result of the decisions in the House of Lords. The equal value issues of all applicants except Mrs Longman were referred to an independent expert and in January 1987 her case was similarly referred. The report was received in February 1988. In June 1988 the Bank amended their Notices of Appearance further to rely upon collective bargaining and arbitration as the material factor defence. In November 1988 the Bank applied for an order that the Applicants should submit themselves for questioning by a second expert appointed by the Bank; this was refused by the Industrial Tribunal and the decision was upheld in this Court: Lloyds Bank Plc v. Fox [1989] ICR 80.In February 1989 the Bank made its concession on equal value and in a joint reply in April of that same year the Applicants alleged discrimination on the S.1(3) issue. The hearing before the Industrial Tribunal took place on various days between 26th June and 20th July 1989. Their decision we have already explained. The Tribunal made a most unusual order adjourning the issue of costs pending the outcome of the present appeal. We do not know the reasons for so doing.Between August 1989 and September 1990 some 750 further originating applications have been lodged by secretaries, typists and clerks employed by the Bank, all claiming equal pay for work of equal value to the work done by such as Mr Martin.

Analytical Job Evaluation Scheme

If part of the reason for bringing these proceedings was to encourage the Bank to initiate a job evaluation scheme then it has been successful. The decision to do so was made in 1988. The Banking, Insurance and Finance Union (BIFU) suggested that the scheme be called the Common Job Evaluation Scheme (CJE). The scheme is based on the Hay Profile Method and bank staff have been trained as evaluators. Evaluating panels are made up of bank staff drawn from those employees covered by the CJE. Half are nominated by the Bank and the other half in equal portions by the two Unions, the Lloyds Group Union (LGU) and BIFU. The LGU, which support the Applicants, still have some reservations about the scheme.The introduction date for the new arrangements was 1st October 1991. On that date 25,634 staff covered by 4,814 jobs had been evaluated. The task was not completed by October 1991 and is continuing. It was anticipated that the evaluation would have been completed by the end of April 1992. There is an ongoing programme of new evaluations and appeals but ultimately approximately 47,500 staff will have been covered by the scheme which will have evaluated 10,952 jobs. We trust by now that this task is completed.It is thus to be hoped that it will be unnecessary in the future for proceedings such as the present to be initiated.The salary of Mr Martin at the relevant time was £8,694. The Applicants salaries were as follows, Mrs Fox - £7.550; Mrs Copley - £6,827; Mrs Laishley - £7,550; Mrs Sissins - £5,386; Mrs Todd - £6,769; Mrs Tofts - £5,863 and Mrs Longman - £6,229.This is said to be the first case before this Appeal Tribunal which involves consideration of the direct effect of sex discrimination in a S.1(3) issue in an equal value claim. The state of the law is such that this Industrial Tribunal faced a formidable task. A number of cases in recent months and years have made clear the extreme complication both in substance and procedures of this branch of the law. Such complications are the enemies of clarity, speed and simplicity of decision. We would respectfully urge statutory reconsideration and clarification as soon as possible.For present purposes we need only refer to direct discrimination in S.1(1)(a) of the 1975 Act ie, it is to treat a woman less favourably than a man "on the grounds of her sex". If the equal pay provisions had formed part of the 1975 Act then the forbidden action might have been stated thus, where a woman or a man is engaged on like work or work rated as equivalent or work of equal value with a man or woman it is unlawful to treat her/him less favourably than her/his comparator by paying her/him less on the grounds of sex. The burden of proving direct discrimination would have been upon the applicant under that Act.The structure of the Equal Pay Act 1970 is however quite different.

Section 1 of the Equal Pay Act 1970 reads as follows -

"1Requirement of equal treatment for men and women in the same employment

(1) If the terms of a contract under which a woman is employed at an establishment in Great Britain do not include (directly or by reference to a collective agreement or otherwise) an equality clause they should deem to include one.(2) An equality clause is a provision which relates to terms (whether concerned with pay or not) of a contract under which a woman is employed (the "woman's contract"), and has the effect that -(a) where the women is employed on like work with a man in the same employment - (i) (ii) ...(b) where the women is employed on work rated as equivalent with that of a man in the same employment - (i) (ii) ... (c)where a woman is employed on work which, not being work in relation to which paragraph (a) or (b) above applies, is, in terms of the demands made on her (for instance under such headings as effort, skill and decision), of equal value to that of a man in the same employment - (i)if (apart from the equality clause) any term of the woman's contract is or becomes less favourable to the woman than a term of a similar kind in the contract under which that man is employed, that term of the woman's contract shall be treated as so modified as not to be less favourable, and(ii)if (apart from the equality clause) at any time the woman's contract does not include a term corresponding to a term benefitting that man included in the contract under which he is employed, the woman's contract shall be treated as including such a term."
"(13) Provisions of this section and sections 2 and 2A below framed with reference to women and their treatment relative to men are to be read as applying equally in a converse case to men and their treatment relative to women."
It must be remembered that sex discrimination is the basis of Article 119 of the Treaty of Rome and the 1970 Act. See ALBION SHIPPING AGENCY V. ARNOLD [1982] ICR 22, 28E: PICKSTONE (supra) at p.715H: LEVERTON (supra) at p.66H. MACARTHYS LTD V. SMITH [1980] ICR 672 , 690 (paragraph 12): JENKINS V. KINGSGATE (CLOTHING PRODUCTIONS) LTD [1981] ICR 592 , 613 (paragraph 10).Each applicant can claim that each and every term of her/his contract of employment should be no less favourable than that of the comparator - HAYWARD v. CAMMEL LAIRD SHIPBUILDERS LTD [1988] ICR 464 .Thus once the concession is made that the jobs in the respective groups of applicant and comparator are of equal value, each applicant, male or female, is entitled to equal pay subject only to a S.1(3) defence, provided always that he or she can find a member of the comparator group who is of the opposite sex.It is clear from PICKSTONE (supra) that the mere fact that the Applicants are engaged on like work with men in their own group does not prevent a claim being brought based on equal value. Thus although in the present case the Applicants happen all to be female, the situation exists here that there could also be male applicants raising similar issues - indeed identical facts.

So far as relevant S.1(3) of the 1970 Act reads -

"An equality clause shall not operate in relation to a variation between a woman's contract and a man's contract if the employer proves that the variation is genuinely due to a material factor which is not the difference of sex and that factor - (a) ... (b) in the case of an equality clause falling within subsection (2)(c) above, may be such a material difference"
It is at this stage that the employer must prove that the difference in pay is not due to sex, and it is clear from some parts of the judgments in the Court of Appeal in PICKSTONE that the fact that there are both men and women on each side of the equation may be relevant in considering a defence under S.1(3). At p.877D, Nicholls LJ says -"If there is a man, or if there are men, doing the same work but being paid no more than a woman, that will be evidence, whose weight will depend upon all the circumstances, that the payment of a higher wage to other men for doing work which is different but of no greater value is due to a material factor other than the difference in sex",

and again at P.883F he says -

"In determining whether the work of these applicants is of equal value to that of the checker warehouse operatives and in determining whether ... the difference in pay between the warehouse operatives and the checker warehouse operatives is explicable by the operation of factors which are unconnected with any discrimination on the grounds of sex, the industrial tribunal should give such weight to the factor that there is a man, or there are men, doing the same work as the applicants and being paid no more, as is appropriate having regard to all the circumstances. These are determinations of fact."
When before the Industrial Tribunal the parties did not agreed upon the correct legal test for direct sex discrimination. The most recent decision at that time and one binding up the Industrial Tribunal was that in JAMES V. EASTLEIGH BOROUGH COUNCIL [1989] ICR 423 - CA. The passage containing the ratio of that decision is contained in the leading judgment of Sir Nicolas Browne-Wilkinson VC at p.428C to 429A which reads -"Mr Lester, for the plaintiff, forcefully submitted that there is direct discrimination in this case. He submitted that discrimination is "on the ground of" sex within section 1(1)(a) if the sex of the plaintiff is a substantial cause of the less favourable treatment. In this context, he says, the correct question is "what would the position have been but for the sex of the plaintiff?" If the position would be different if the plaintiff's sex were different, that is direct discrimination.I do not accept that construction of section 1. In my judgment section 1(1)(a) is looking to the case where, subjectively, the defendant has treated the plaintiff less favourably because of his or her sex. What is relevant is the defendant's reason for doing an act, not the causative effect of the act done by the defendant. As Mr Towler for the council pointed out, section 1(1) is referring throughout to the activities of the alleged discriminator. In the case of direct discrimination "a person discriminates against a [man] ... if on the ground of [his] sex he treats [him] less favourably. The relevant question is "did the defendant act on the ground of sex?" not "did the less favourable treatment result from the defendant's actions?" Thus, if the overt basis for affording less favourable treatment was sex (e.g. an employer saying "no women employees") that is direct discrimination. If the overt reason does not in terms relate to sex (e.g. in selection for redundancy, part-time employees are the first to go) that is not on the face of it direct discrimination since sex does not come into the overt reason given for the action. If, but only if, it is shows that the overt reason is not the true reason but there is a covert reason why the employer adopted those criteria (e.g. to get rid of his female employees) will it be direct discrimination. In such a case the true reason for the policy is the desire to treat women less favourably than men: the employer is therefore acting on that ground.But in the case where neither the overt condition imposed nor any covert reason relates directly to the sex of the plaintiff, in my judgment it cannot be said that the defendant afforded less favourable treatment "on the ground" of sex. He acted not on the ground of sex but on other grounds. The result of his acting on grounds other than sex may produce a disparate impact on men and women: if so his actions may constitute unlawful indirect discrimination. He has not acted on the ground of the plaintiff's sex, since that was not his reason for adopting the policy he has adopted."It is also to be noted that the Learned Vice-Chancellor distinguishes the decision in the House of Lords in REG V. BIRMINGHAM CITY COUNCIL, EX PARTE EQUAL OPPORTUNITIES COMMISSION [1989] AC 1155 .As we read the written submissions, Mr Lester repeated the argument raised before the Court of Appeal in this hearing before the Industrial Tribunal and the Bank was arguing that the Tribunal was bound by JAMES in the Court of Appeal.That decision in the Court of Appeal was reversed by the House of Lords at [1990] 2 AC 751 . The appropriate test now therefore is that which was being advanced for the Equal Opportunities Commission in the Court of Appeal in JAMES , namely would the applicant suffer the alleged detriment or less favourable treatment, but for being a woman or a man. Thus in considering the Applicants Reply alleging direct discrimination, it is to allege that but for being women they would at the time of their applications be receiving the same pay as their male comparator.In the present case - bearing in mind that each case is an individual application - the question can be put "In February 1986 (the date of the Originating Application) would the applicant have received the same salary as Mr Martin if she had been a man - but for being a woman?"; or "Would a man in the applicant's position have received the same salary as Mr Martin?". In posing these questions one must not lose sight of the burden of proof. It is for the employer to prove that the difference in pay is not due to sex. Put another way, that the Applicant would still be receiving the same pay even if she had been a man.There is no issue here but that the material factor alleged and to be considered is the existence of collective bargaining and arbitration. It must be genuine and causative. It must also be untainted by discrimination on the grounds of sex. As we understand it the real issue between the parties was on that last point. The Applicants allege both direct and indirect discrimination. The allegation of direct discrimination was in effect "stereotyping". Thus considering direct discrimination, if we have understood correctly the principle of JAMES to be applied in the context of S.1(3) of the 1970 Act, there is the need to compare the position of an applicant with someone of the opposite sex in the same position and then possibly to ask whether he or she is worse off because of sex. The burden upon the employer is to show that he or she is not.

The Decision of the Industrial Tribunal

In paragraphs 1 - 5 the Tribunal set out the issues. In paragraphs 6 - 15 they describe the negotiating basis. In paragraphs 16 - 19, the evidence. In paragraphs 20 - 32 they consider direct discrimination. Indirect discrimination is considered in paragraphs 33 - 37 and the conclusion is contained in paragraph 38. They find as follow -
"For the above reasons we find that the applicants submissions in relation to indirect discrimination must be rejected. We reiterate that our finding is that the bargaining process on which the respondent relies as a material factor was tainted by direct sex discrimination and cannot therefore constitute an offence under S.1(3)"
Nowhere in the decision does the Tribunal direct itself on the correct question of law to be asked when considering direct sex discrimination. Differing submissions on that legal issue had been presented to it. It was essential for the parties, and is essential for us, to know the legal basis upon which their decision was reached. The Tribunal was bound by JAMES in the Court of Appeal and should have applied the principles therein set out; if it had done so, it would have erred in law. Can we be certain that it did not? We are unable to discover which principle was applied and we are therefore driven to the conclusion that as the basis of the legal approach is left uncertain and indeed undecided, this decision must be treated as flawed.We therefore turn to the decision in DOBIE V. BYRNES INTERNATIONAL SECURITIES SERVICES (UK) LTD [1984] ICR 812 and ask ourselves, having found a legal error in the decision, whether the decision is plainly and unarguably right notwithstanding that error. It is only if it is plainly and unarguably right notwithstanding that error that the decision can stand. If we conclude that the decision was wrong or might have been wrong then it is for us to remit the matter for a rehearing.We therefore turn to consider the findings of fact upon which this decision is based and whether that decision can be considered to be "plainly and unarguably right notwithstanding that error".The negotiating arrangements have varied for each of the relevant groups of staff.The messengers are classified as "T & S" (technical and services) staff. Since 1970 the basic grade messengers have been the subject of national collective bargaining in the Joint Negotiating Council for non-Clerical Banking Staff. The Employers were represented by the Federation of London Clearing Bank Employers and until 1980 the staff were represented by the Banking Staff Council for non-Clerical Staff which was an amalgam of various associations and trade unions. From 1981 two Trade Unions have represented the staff, CBU and BIFU. These Trade Unions insist on separate negotiations with each. Those negotiations were normally held on the same day and no difference in the results obtained.Arbitrations on pay claims occurred in 1970 and 1979. In 1980 there was a specific incident which affected pay. The National Westminster Bank at that time operated outside the Federation and as a result of industrial action agreed a substantial increase in messengers' pay. The Federation felt it necessary to bring messengers' pay into line with that achieved at the National Westminster Bank and the flat-rate increase of £212 in basic pay per annum was granted. Subsequently the National Westminster Bank rejoined the Federation in relation to non-clerical staff.Pay for the clerk-typists was negotiated domestically before 1968. From 1968 to 1977 the rate of pay to the national maximum had been the subject of national bargaining. The parties were the Federation and the Banking Staff Council for clerical staff. After a break of three years national negotiations were resumed in 1980 the staff being represented by CBU and BIFU. Here again the trade unions insist on separate negotiations. There was an arbitration in 1972 and since 1970 the rates have been the result of collective bargaining.Secretaries' and typists' pay has always been the subject of domestic negotiation and not of national collective bargaining. The bargaining is through the Joint Domestic Negotiating Council (clerical). From 1971 to 1979 staff was represented jointly by BIFU and the Lloyds Bank Group Staff Association. However since 1980 when the Staff Association became part of CBU there has been insistence on separate negotiations, it would seem that pay has been strongly influenced by the results of collective bargaining by the Joint Negotiating Council for Clerical Staff which is a national bargaining body.The evidence is summarised in paragraphs 16, 17, 18 and 19 of the Decision. There was a vast mass of written material consisting in the main of contemporaneous minutes of meetings of the various national and domestic negotiating bodies going back in some cases over 20 years. There were also internal records of the Federation. Three witnesses were called for the Bank. Mrs Marsland (NUBE 1968-1972) was the sole witness for the Applicants. As the Tribunal commented, " For the most part, the witnesses were unable to give first hand evidence on these matters and the Tribunal has had to form its own view on the meaning and significance of the minutes ".The Tribunal were critical of the value of those minutes and the Bank in its submissions before us described the criticisms as indicating that the minutes were a "charade". The Applicants make no attempt to support such a comment and in the main the Applicants are content to argue that the minutes do not really assist the matter one way or the other. The Applicants were contending that the negotiating machinery by which the results of pay were achieved was not the cause of the resulting pay awards. Their case was summarised by the Tribunal at paragraph 21 as follows -
"They say that the cause of the pay differences here was (a) the traditional view and assumptions on all sides about the value (in money terms) of the respective jobs; (b) similar traditional assumptions about men being breadwinners and women being secondary wage earners; and (c) the respondent's failure at any time before 6 February 1989 to accept that the jobs were of equal value."
It must not be forgotten that discrimination on the grounds of sex was permitted until December 1975 and there would have been no reason to avoid reference to those matters in any documentary records. Moreover it was not until after the Equal Pay (Amendments) Regulations 1983 came into effect that an equal value claim could be brought in this country.Before turning to the more detailed analysis of the reasoning and findings of this Tribunal, we would only comment that we doubt whether much assistance can be derived from general reports of a particular sector in the industrial world. The Tribunal clearly relied upon a report of the Equal Opportunities Commission on Barclays Bank. It must not be forgotten that the Tribunal is dealing with a particular applicant and a particular bank and that is the issue which the Tribunal must decide. This Bank had no opportunity of challenging any of the evidence upon which that report was based nor the reasoning upon which the conclusions were drawn.The two most important paragraphs in the Decision are paragraphs 23 and 28. They read as follows:-
"23 It is common ground that the applicants are paid less than the male comparator for work of equal value. The secretaries, typists and clerk typists employed by the respondent consist mainly of women. The messengers consist almost entirely of men and originally consisted entirely of men. Against that background the onus is on the respondent to prove an objective non-sex based reason for the differences in pay between the two groups. Having regard to the comments we have made in paragraph 18 on the documentary evidence, it appears to us that the respondent is in a real difficulty here. Even in the absence of any evidence from the applicants, we should find it hard to believe that the traditional assumptions referred to in paragraph 21 were wholly absent from the minds of the negotiators and the ultimate arbiters of at least some of the pay awards made over the years. It seems to us highly probable that such assumptions did influence such awards, whatever the minutes may or may not say. To the extent that that happened, whether on a conscious or an unconscious level, there is a taint of direct discrimination and that, we find, is insufficient to refute the respondent's argument that the material factor is not the difference of sex."
24 - 27 ...
"28 To sum up so far, we find that the pay inequality complained of originated in the early 1970s on the basis of gender based assumptions and attitudes; that the pay inequality remained and increased: see Mr Waine's proof of evidence, paragraphs 35 to 47, and the statistics he produced (Schedules A, B and C). Further, there is no evidence that the respondent sought to eliminate sex discrimination as between the respective pay structures so as to remove the inequality complained of. Thus, even if the respondent could prove that each individual pay award over the years was wholly free of the taint of sex discrimination, the already existing taint was not removed, so that argument would be of no avail."
The first criticism of these paragraphs made by the Bank is that the Tribunal was basing its decision on assumptions and those were the assumptions put forward by the Applicants in paragraph 21.It has been clearly recognised that an Industrial Tribunal is an industrial jury and in that capacity it can bring to its deliberations the industrial experience of the lay members. This must not be allowed to cloud the duty of a Tribunal to find facts based upon evidence not only primary facts, but reasoned and reasonable inferences to be drawn from those facts. We considered the limitations placed upon lay members recently in HALFORD v. SHARPLES & ORS [1992] ICR 146.In the present case the Industrial Tribunal has purported to make findings of facts based upon assumptions, which were not based upon evidence. The Tribunal has seemingly started from its own basis of fact without so declaring it. Indeed, it would be almost impossible, say the Bank, for a party to refute such an assumption and this is emphasised by the comment in paragraph 23 that, "... the respondent is in a real difficulty here." Moreover, say the Bank, the chances of defeating a finding by an Industrial Tribunal of an unconscious assumption are indeed remote.The detailed submissions of the Bank on this point were far more extensive than we have indicated and were impressive. We hope that we have not been too succinct in our acceptance of them.We therefore find that this decision is flawed in its findings of fact. Such assumptions seem to have coloured the whole approach of the Tribunal. This is particularly clear when set against the massive documentary evidence as to which in paragraph 27 of the decision it finds that despite the submission of the Applicants there is no evidence of direct discrimination from that documentation. The Tribunal has cast aside the massive evidence of that documentation which was contemporaneous, subject to detailed criticism and amendment before agreement, and agreed at a time when there was no reason in terms of sex discrimination for the contents to be other than frank and open reports of negotiations.The second criticism made of paragraph 23 is that the first three sentences reflect considerations applicable to indirect discrimination and thus in the absence of a clear legal direction on direct discrimination, show confusion of approach. We agree. This merely emphasises the importance of the absence of a self-direction on the law to be applied.A similar comment could validly be made of the fourth sentence.The assumptions put forward are set out in paragraph 21 as follows.
"... They say that the cause of the pay differences here was (a) the traditional view and assumptions on all sides about the value (in money terms) of the respective jobs; (b) similar traditional assumptions about men being breadwinners and women being secondary wage earners; and (c) the respondent's failure at any time before 6 February 1989 to accept that the jobs were of equal value."
We are unable to accept that that third assumption points in any way towards direct discrimination in 1986. An equal value claim was indicated only shortly before that and a party must surely be allowed to dispute such a claim without an assumption being made that such action, of itself, amounts to direct sex discrimination. An assessment of equal value was only relevant from early 1984.The Bank argue that it is difficult to understand how the first assumption, even if valid, can be related to direct sex discrimination in pay in 1986 or indeed to any sex discrimination. As we understand this, it is the same point as is made in paragraph 31 of the decision where the Tribunal say this -
"This pattern of inequality derives partly from the under-valuation by many employers of work in predominantly female occupations and from the weaker bargaining collective bargaining power in such occupations. Both these factors apply in the present case. No effort has been made by the respondent to deal with this pattern of inequality."
Some would refer to this issue as a matter of "stereotyping", namely that because the work is done by a woman it is less valuable. It would be rated more highly if done by a man. There is a failure therefore to look at the job itself in its assessment. However, a very careful analysis is required and one must ask whether the diminution in pay is attributable to the job itself and/or to facts known about the comparative groups, or is it not due not to those factors but to the fact that because the applicant group is female the difference is attributable to the fact that the work is being done by women. The relevant date for these applications is 1986.In the present case a significant number of clerical workers were male and spread across the grades. Equalisation had occurred and women clerical staff received male rates. What if the Applicants in this case had been from the male clerks? Could a male applicant from these same groups as the appellants, rely upon that same evidence? If not, a female applicant could succeed on an equal value clause where a male applicant could not. Presumably he would, after her success, have to claim against another female applicant in his own group on a like work basis.The use of the word "traditional" can, in some instances nowadays, be rather misleading as so many jobs are now being carried out both by men and by women alike. It will be some measure of success in the elimination of discrimination if that word "traditional" can be rendered otiose.Whilst not taking so clear a view of the second stated assumption, without a very clear and detailed analysis of approach we have doubts that it is relevant to direct sex discrimination. We were presented with extensive submissions from both sides upon what was termed this "breadwinners assumption". It is unnecessary for us to set out and comment on these arguments save to say that some of us accept the logical force of the argument presented in the Bank's reply at p.25 et seq.However, even if it were a valid finding of fact based upon evidence not just an assumption, the Bank is nevertheless left in the position that there is a finding against it of direct discrimination based upon three assumptions of sex discrimination at least two of which are invalid as such.This of itself must, in our judgement, render a decision unsafe and unsatisfactory.In the light of these comments already made, we do not find it necessary to recite and review the many criticisms made by the Bank at the approach to documentary evidence made in paragraph 18 of the decision. We were impressed by them. It is important to remember that the witnesses in the main were not able to speak of their own knowledge. Some were involved with negotiation for only a very short period.

The reasoning in paragraph 28 of the decision seems to run thus:-

1. There was sex discrimination "in the early 1970s".The comparative groups obtained broadly similar percentage increases between that time and 1986 and hence differentials remained fairly constant.The Bank took no positive steps to eradicate that sex discrimination.Therefore even if the pay awards between 1970 and 1986 were negotiated without any element of sex discrimination, nevertheless the original sex discrimination continued.It was suggested during submissions that discrimination "in the early 1970s" was not based upon an assumption, but upon the complex restructuring of clerical staff in 1971/72. This had indeed been argued before the Tribunal at length and had it intended to rely upon any such evidence, it would undoubtedly and should undoubtedly, have made a specific finding. We were referred to the relevant evidence and most, if not all, of the members of this Court are satisfied that there was no discrimination in restructuring of clerical staff. Mrs Marsden was unable to give first-hand knowledge about it. Thus, this finding of the Industrial Tribunal was based upon an assumption.Even if this assumption was accurate, it is difficult to understand how it can be argued that the discrimination continued until 1986. "Broadly speaking, percentage differentials were maintained" (paragraph 26). This does not deal with the causes of pay in 1986. In each year pay was re-negotiated; hence each year was a fresh award. In the interim, pay between males and females had been equalised. Arbitrators awarded changed benefits and remuneration. Government pay freezes had varying effects between groups. In 1980 there was an exceptional award resulting from industrial action.The numerical make up of the trade unions were broadly equal between male and female. Negotiators included males and females. The "industrial muscle" of the clerical staff must surely have been considerable. Industrial action by them would surely be as effective to interfere with the banking process as that of messengers. Industrial action by the clerical staff had in fact taken place since 1972.The issue of differentials seems to us to be an unsafe basis for criticism of direct sex discrimination. Even if there had been evidence which clearly identified specific sex discrimination in 1970, the question is whether it was still a cause in 1986, not whether differentials remain broadly the same.Suppose for instance there was sex discrimination between group 'A' and group 'B' in 1970 and that between the years 1970 and 1986, the pay of group 'A' had been linked to a wholly different group 'C' and that of group 'B' to a different group 'D', then differentials might be the same but the cause, the reason for the differential would cease to be that of the difference of sex. This is but one example, others come readily to mind.Clearly therefore a detailed examination year-by-year is necessary if accurate tracing of the initial cause is to be achieved.We have found no evidence which could justify such a finding in the present case. That is not to say that a specific discriminatory factor could not be found still to exist at a later date.The huge, extensive and undoubtedly expensive historical retracing and examination of wage negotiations over something like 20 years has caused us anxiety. We would hope that as the y ears pass by it will be thought unnecessary for such an investigation. The issue in direct sex discrimination is clear and we would also have thought that where indirect sex discrimination is involved the requirement or condition could be easily identified and that the evidence could be somewhat restricted.In paragraph 29 of the decision, the Industrial Tribunal refers to recruitment of messengers, but does not refer to direct discrimination. We do not understand that it was suggested that this affected pay. Promotion within the clerical grades is not suggested to be discriminatory. It was not argued that this affected pay. Neither of these matters indicate to us any direct sex discrimination.Having thus examined the reasoning behind this decision and the basis upon which it was reached, we, the majority, feel ourselves unable to accept those findings or that reasoning. Thus applying Dobie we must remit this case to be reheard before a different Industrial Tribunal upon the issue whether the Bank has proved a defence under S.1(3) so far as it has to meet the allegation by the Applicants of direct discrimination.The dissenting member (Ms Smith) accepts the Industrial Tribunal's conclusion that there was direct discrimination against the Applicants. The Applicants' work was found by an independent expert to be of equal value to that of a senior messenger. Three years after the Applicants' original claim that their work was of equal value, and following the report of the independent expert, the bank conceded that their work was of equal value with one of the comparators, Mr Martin, a senior messenger.Before the Industrial Tribunal Lloyds relied on S.1(3) of the Equal Pay Act 1970 - that the variations in pay were due to a material factor which was not the difference of sex. The Tribunal, sitting for seventeen days, decided that the bank had not made out a defence to the claim. The onus was upon the bank to satisfy the Tribunal. They failed to do so.There was evidence upon which the Tribunal reached their conclusion. There was the evidence of Mrs Marsland which the Tribunal said, in their paragraph 25, they accepted "without reservation". There was the evidence of Mr Waine and Mr Cullam in cross-examination.She accepts the reasoning of the Tribunal in their paragraph 21 that:
"... the cause of the pay difference here was (a) the traditional view and assumptions on all sides about the value (in money terms) of the respective jobs; (b) similar traditional assumptions about men being bread-winners and women being secondary wage earners; and (c) the respondent's failure at any time before 6th February 1989 to accept that the jobs were of equal value."
In the absence of an acceptable explanation it was logical for the Industrial Tribunal to look at the history of the pay discrepancies.Paragraphs 23 and 28 of the Tribunal's findings, cited above, should be taken in context with the accompanying paragraphs 24, 25, 26 and 27. They read:
"24. The applicants go further and produce evidence. The most significant evidence, we find, is that of Mrs Marsland, who said that during her time as a national officer with NUBE, between 1968 and 1972, there were both clear and recognised assumptions by both parties to the collective bargaining that in negotiations for messengers they were dealing with 'men's work'. For example, the Beveridge notion of a social or family wage always figures highly in the negotiations. She was confident that this was so, whether or not it was reflected in the minutes. This was because messengers joining the respondent bank at an early age were married and had families to support. She also said that when they were looking for groups to compare messengers with, they chose the manual sections of industry, which were predominantly male.
25. Mrs Marsland recalled that, in the early 1970s, there was a restructuring of clerical grades and salaries in part in preparation for equal pay legislation. However, the bank never discussed with her union the proposals for action to comply with the Equal Pay Act. It was clearly understood that messengers' work was men's work. At that time the legislation gave redress only where there was like work or broadly similar work. Since no women were doing messengers' work, the legislation was regarded as irrelevant. We accept Mrs Marsland's evidence without reservation.26. Mr Waine also accepted in cross-examination that the pattern of pay increases remained fairly consistent over time, with similar percentage increases for clerical and messenger staff. Broadly speaking, percentage differentials were retained. Mr Richards stated in cross-examination that it would be predominant in the thinking of bank people that the messenger was a family man who needed to be paid a family wage.27. The applicants pointed to certain passages in the minutes which, they said, indicated gender based attitudes and assumptions. Although they may be straws in the wind we do not find them to be proof of direct sex discrimination. Here the applicants are in the same difficulty as the respondent; the difficulty of provind the actual cause of any particular pay award."Ms Smith thinks that the attitude of the Tribunal to the minutes of the various negotiating bodies was a realistic one. It did not regard the minutes as a 'charade'. But industrial relations practitioners would not expect to see open references to pay awards based on sex even though sex discrimination was lawful until December 1975. For many years before the legislation there was a sustained campaign against sex discrimination and it was that campaign which helped to lead to the introduction of the legislation. Overt references to gender based pay awards before the end of 1975 would have been lawful but they would have been very unpopular with considerable sections of the workforce and both sides of industry were conscious of that.Reliance on such arguments as extra pay for "breadwinners" were not uncommon. Ms Smith does not accept the bank's argument that it is not sexually discriminatory to apply a particular benefit to a group of men (or women) where it is based on "known facts". She feels that in order not to discriminate the employer must ascertain whether there are employees in the other gender group with similar family responsibilities - ie single parents.Ms Smith agrees with the tribunal that national pay bargaining can constitute a material factor defence but that in this instance the pay bargaining was tainted by sex discrimination which originated in the earlier pay structure and was not eliminated. Moreover, arbitration awards are unlikely to strike new ground unless the arbitrators are asked so to do. Arbitrators work within the parameters presented to them. They are used to solve a particular industrial problem and their decision is accepted, sometimes reluctantly, in that climate.Ms Smith feels that the Tribunal were conscious of the causal test under James v. Eastleigh Borough Council and that they decided that the applicants suffered less favourable treatment because they were women. There was, in their view, direct discrimination. She believes that the Tribunal applied the law and that there was no factual findings upon which the EAT has grounds to intervene. There was evidence upon which they reached their conclusions and their decision should be upheld.Because of the view of the majority it must follow that the appeal is therefore allowed and the cross-appeal on indirect discrimination is dismissed.Leave to appeal on the appeal and cross-appeal.